These functions determine which items in a vector can be considered (the start of) a new episode, based on the argument episode_days. This can be used to determine clinical episodes for any epidemiological analysis. The get_episode() function returns the index number of the episode per group, while the is_new_episode() function returns values TRUE/FALSE to indicate whether an item in a vector is the start of a new episode.
Arguments
- x
vector of dates (class
DateorPOSIXt), will be sorted internally to determine episodes- episode_days
required episode length in days, can also be less than a day or
Inf, see Details- ...
ignored, only in place to allow future extensions
Details
Dates are first sorted from old to new. The oldest date will mark the start of the first episode. After this date, the next date will be marked that is at least episode_days days later than the start of the first episode. From that second marked date on, the next date will be marked that is at least episode_days days later than the start of the second episode which will be the start of the third episode, and so on. Before the vector is being returned, the original order will be restored.
The first_isolate() function is a wrapper around the is_new_episode() function, but is more efficient for data sets containing microorganism codes or names and allows for different isolate selection methods.
The dplyr package is not required for these functions to work, but these functions do support variable grouping and work conveniently inside dplyr verbs such as filter(), mutate() and summarise().
Examples
# `example_isolates` is a data set available in the AMR package.
# See ?example_isolates
df <- example_isolates[sample(seq_len(2000), size = 200), ]
get_episode(df$date, episode_days = 60) # indices
#> [1] 61 24 5 23 35 13 33 39 24 47 9 17 54 13 24 5 22 56 7 12 2 47 26 33 35
#> [26] 7 23 6 49 45 58 31 1 2 2 11 61 40 27 59 31 2 61 20 1 20 14 4 33 28
#> [51] 31 50 51 5 44 38 61 43 15 6 52 13 46 16 16 21 23 22 16 32 5 12 35 57 6
#> [76] 2 61 21 17 9 57 10 6 43 60 28 38 34 56 27 28 37 54 28 56 12 41 7 17 44
#> [101] 35 6 45 50 52 57 25 34 13 1 40 43 61 48 2 49 38 50 31 8 61 47 10 3 9
#> [126] 12 37 11 44 35 19 56 37 37 38 10 51 4 9 21 45 16 52 26 23 25 30 42 46 57
#> [151] 10 29 29 23 6 21 50 38 11 18 10 11 30 13 57 50 39 10 10 32 53 51 28 52 55
#> [176] 59 49 13 58 17 4 16 37 17 10 47 32 11 14 60 28 13 55 36 14 10 26 54 32 8
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE FALSE TRUE FALSE FALSE
#> [13] FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [25] FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE
#> [37] FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE TRUE
#> [49] FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE TRUE FALSE TRUE FALSE
#> [61] FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE
#> [73] FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE
#> [85] TRUE FALSE FALSE TRUE TRUE TRUE FALSE TRUE TRUE FALSE FALSE TRUE
#> [97] TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [109] FALSE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE
#> [121] FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE TRUE FALSE TRUE FALSE
#> [133] FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [145] FALSE TRUE TRUE TRUE TRUE FALSE FALSE FALSE TRUE FALSE TRUE TRUE
#> [157] TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE
#> [169] FALSE FALSE TRUE FALSE TRUE FALSE TRUE TRUE FALSE FALSE TRUE FALSE
#> [181] FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE
#> [193] FALSE TRUE FALSE FALSE TRUE FALSE FALSE TRUE
# filter on results from the third 60-day episode only, using base R
df[which(get_episode(df$date, 60) == 3), ]
#> # A tibble: 1 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-05-16 D25302 65 F ICU B_STRPT_ANGN S NA NA S
#> # … with 36 more variables: AMC <rsi>, AMP <rsi>, TZP <rsi>, CZO <rsi>,
#> # FEP <rsi>, CXM <rsi>, FOX <rsi>, CTX <rsi>, CAZ <rsi>, CRO <rsi>,
#> # GEN <rsi>, TOB <rsi>, AMK <rsi>, KAN <rsi>, TMP <rsi>, SXT <rsi>,
#> # NIT <rsi>, FOS <rsi>, LNZ <rsi>, CIP <rsi>, MFX <rsi>, VAN <rsi>,
#> # TEC <rsi>, TCY <rsi>, TGC <rsi>, DOX <rsi>, ERY <rsi>, CLI <rsi>,
#> # AZM <rsi>, IPM <rsi>, MEM <rsi>, MTR <rsi>, CHL <rsi>, COL <rsi>,
#> # MUP <rsi>, RIF <rsi>
# the functions also work for less than a day, e.g. to include one per hour:
get_episode(c(
Sys.time(),
Sys.time() + 60 * 60
),
episode_days = 1 / 24
)
#> [1] 1 2
# \donttest{
if (require("dplyr")) {
# is_new_episode() can also be used in dplyr verbs to determine patient
# episodes based on any (combination of) grouping variables:
df %>%
mutate(condition = sample(
x = c("A", "B", "C"),
size = 200,
replace = TRUE
)) %>%
group_by(condition) %>%
mutate(new_episode = is_new_episode(date, 365)) %>%
select(patient, date, condition, new_episode)
}
#> # A tibble: 200 × 4
#> # Groups: condition [3]
#> patient date condition new_episode
#> <chr> <date> <chr> <lgl>
#> 1 5DB1C8 2017-12-28 C FALSE
#> 2 921720 2007-11-02 A FALSE
#> 3 088256 2003-01-25 B TRUE
#> 4 122506 2007-08-11 B FALSE
#> 5 E58716 2010-07-29 A TRUE
#> 6 753036 2004-11-15 B FALSE
#> 7 4F6830 2010-03-12 C FALSE
#> 8 687590 2011-10-29 C FALSE
#> 9 965996 2007-12-03 A FALSE
#> 10 778835 2014-03-24 A FALSE
#> # … with 190 more rows
if (require("dplyr")) {
df %>%
group_by(ward, patient) %>%
transmute(date,
patient,
new_index = get_episode(date, 60),
new_logical = is_new_episode(date, 60)
)
}
#> # A tibble: 200 × 5
#> # Groups: ward, patient [186]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Clinical 2017-12-28 5DB1C8 1 TRUE
#> 2 Clinical 2007-11-02 921720 1 TRUE
#> 3 ICU 2003-01-25 088256 1 TRUE
#> 4 Clinical 2007-08-11 122506 1 TRUE
#> 5 Clinical 2010-07-29 E58716 1 TRUE
#> 6 Clinical 2004-11-15 753036 1 TRUE
#> 7 Clinical 2010-03-12 4F6830 1 TRUE
#> 8 Clinical 2011-10-29 687590 1 TRUE
#> 9 Clinical 2007-12-03 965996 1 TRUE
#> 10 Clinical 2014-03-24 778835 1 TRUE
#> # … with 190 more rows
if (require("dplyr")) {
df %>%
group_by(ward) %>%
summarise(
n_patients = n_distinct(patient),
n_episodes_365 = sum(is_new_episode(date, episode_days = 365)),
n_episodes_60 = sum(is_new_episode(date, episode_days = 60)),
n_episodes_30 = sum(is_new_episode(date, episode_days = 30))
)
}
#> # A tibble: 3 × 5
#> ward n_patients n_episodes_365 n_episodes_60 n_episodes_30
#> <chr> <int> <int> <int> <int>
#> 1 Clinical 115 14 56 78
#> 2 ICU 61 12 35 45
#> 3 Outpatient 10 7 9 9
if (require("dplyr")) {
# grouping on patients and microorganisms leads to the same
# results as first_isolate() when using 'episode-based':
x <- df %>%
filter_first_isolate(
include_unknown = TRUE,
method = "episode-based"
)
y <- df %>%
group_by(patient, mo) %>%
filter(is_new_episode(date, 365)) %>%
ungroup()
identical(x, y)
}
#> Including isolates from ICU.
#> [1] FALSE
if (require("dplyr")) {
# but is_new_episode() has a lot more flexibility than first_isolate(),
# since you can now group on anything that seems relevant:
df %>%
group_by(patient, mo, ward) %>%
mutate(flag_episode = is_new_episode(date, 365)) %>%
select(group_vars(.), flag_episode)
}
#> # A tibble: 200 × 4
#> # Groups: patient, mo, ward [196]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 5DB1C8 B_STPHY_EPDR Clinical TRUE
#> 2 921720 B_STPHY_CONS Clinical TRUE
#> 3 088256 B_STPHY_CONS ICU TRUE
#> 4 122506 B_STPHY_AURS Clinical TRUE
#> 5 E58716 B_STPHY_EPDR Clinical TRUE
#> 6 753036 B_STPHY_EPDR Clinical TRUE
#> 7 4F6830 B_ESCHR_COLI Clinical TRUE
#> 8 687590 B_STPHY_CONS Clinical TRUE
#> 9 965996 B_STRPT_PNMN Clinical TRUE
#> 10 778835 B_STPHY_AURS Clinical TRUE
#> # … with 190 more rows
# }